Overheating at surface and through water column

dynamics-numerics-params
, ,
1

Hi there,

I have set up a 1km scale model and am having difficulty with overheating in the surface layer and throughout the water column. This heating up occurs within the first day and remains. I am using bulk fluxes. I have tried with both hourly ERA5 (from cds store using ERA5 scripts) and and the default bulk fluxes built into crocotools, but I get the same effect with both. I have attached my cpp file for inspection cppdefs.h;
I am currently working through a few possibilities related to vertical mixing parametrisation and OBC options, but nonetheless I would appreciate any suggestions based on your experience or knowledge.

Thanks in advance,
Joe

UPDATE: I am still asking for your suggestions and experience, but it seems (provisionally) that reverting to v1.1 solves this issue. I had switched everything to v1.2 (then v1.2.1) and it seems that there is some major difference, which either requires another CPP flag that isn’t contained in the default cppdefs.h, or there is something buried in the code which causes issues in my model domain, but not for other models used for benchmarking.

Here is the cppdefs.h file for v1.1 cppdefs.h

2

Hello,
Your issue is still there with the v1.2.1 version ?
You say it’s okay in v1.1, could you post your cppdefs.h from this version please ?

did you try to use the cpp key define BULK_LW ?

Matthieu

3

Hi Matthieu,

I’ve attached the cppdefs.h from v1.1 in the original post above.

When using BULK_FLUXES, I think I recall using BULK_LW. I can try again and check in case though.

Thanks,
Joe

4

from your cppdefs.h in v1.1 you didn’t use BULK_FLUX (it’s set to UNDEF).

5

That’s correct. I have since used it and it works fine.

6

Hi jmcgover,
Do you mean it is now working fine with BULK_FLUX with the v1.2.1 croco latest version ?
Gildas

7

Hi GildasC, the issue is resolved when I revert to v1.1 but I haven’t done any more testing with v1.2.1.